Reproduce “An Improved Autoencoder Approach for Nuclei Image Segmentation” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Aydin Ayanzadeh, Shokoufeh Yazdanian
The authors introduce a U-Net for pathology nuclei segmentation that uses residual squeeze-and-excitation blocks in skip connections and a densely connected squeeze-and-excitation block at the bottleneck. On the 2018 Data Science Bowl task, Table 1 reports 94.13% Dice and 87.92% IoU, versus U-Net’s 90.31% and 82.49%. Conflicting scores elsewhere in the paper prevent a reliable ranking or causal conclusion.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.